Although many studies have established that adherence to HIV medication is closely associated with viral suppression, drug resistance, disease progression and death, most data were collected on obsolete, partially suppressive regimens with coarse patient-reported adherence measures, limited sample size and statistical power. Investigations using objective measures have examined only traditional raw percent adherence and have not yet addressed other aspects of adherence behavior such as dose timing. As such, critical questions remain unanswered. For example, which single antiretroviral medication (ARV) is associated with the best adherence, how adherence behavior determines whether patients develop either limited or multi-drug resistance, and how complex adherence behavior, including that not captured by percent adherence, influences virologic and clinical outcomes. In addition, most studies that utilized traditional adherence measures only examined viral suppression as the primary outcome, but not other clinically important outcomes such as CD4 cell loss, disease progression and death. Because objective adherence measures, such as electronic monitoring systems, are expensive, most large studies have been limited to self-report or pharmacy refill, which are limited by measurement bias, lack of precision and inability to capture all aspects of adherence behavior in order to objectively describe their impact on outcomes. We propose a multi-site collaborative study that integrates adherence with wide coverage of ARVs and regimens, demographics, psychosocial and behavioral characteristics, and virologic and clinical outcomes data from 3,003 individuals across 15 carefully selected studies. This multi-site collaboration study will: (1) create a rich and powerful data source, coordinated by a Statistical Coordination and Analysis Center, for a broad range of investigators to address important questions that are difficult or impossible to answer with any single data set; (2) create and validate adherence measures that can characterize multi- faceted adherence behavior fully, objectively and accurately; (3) study patterns and determinants of adherence to ARVs, as well as evaluate the effectiveness of interventions with large, diverse HIV population; and (4) model virologic outcomes, clinical events, disease progression and death as functions of adherence to ARVs, patient individuality, medication and regimen characteristics and other factors over time. [unreadable] [unreadable] [unreadable]